Social networks are complex systems in which individuals form evolving communities. Traditional community detection methods struggle with these changes, leading to inaccurate results. This research develops new algorithms to identify and track dynamic communities accurately. The proposed algorithms adapt to social network changes, providing more accurate community structures over time. A rigorous evaluation of real-world datasets demonstrated their effectiveness and efficiency. The proposed algorithm focuses on the temporal changes in the network over time, and by using the modularity graph metric, we can understand the connectivity between the communities in the network. Initially, the static algorithm might provide better results since it quickly adapts to new data. However, over time, the dynamic algorithm offers a greater accuracy, stability, and generalized modularity results as it considers all the historical data. A comparative analysis was conducted between the static modularity results and the dynamic modularity results, incorporating data from the previous 50 occurrences. The findings indicate that the dynamic approach yields superior results. Additionally, extending the analysis to include all previous data provided the most stable and accurate outcomes.